diff options
Diffstat (limited to 'ggml-cuda/common.cuh')
| -rw-r--r-- | ggml-cuda/common.cuh | 4 |
1 files changed, 4 insertions, 0 deletions
diff --git a/ggml-cuda/common.cuh b/ggml-cuda/common.cuh index 44e67e04..b6f0bc36 100644 --- a/ggml-cuda/common.cuh +++ b/ggml-cuda/common.cuh @@ -321,6 +321,10 @@ static __device__ __forceinline__ int __dp4a(const int a, const int b, int c) { #define FP16_MMA_AVAILABLE !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__)) && __CUDA_ARCH__ >= CC_VOLTA +static bool fast_fp16_available(const int cc) { + return cc >= CC_PASCAL && cc != 610; +} + static bool fp16_mma_available(const int cc) { return cc < CC_OFFSET_AMD && cc >= CC_VOLTA; } |